CN108700854A - For according to control module come the method and apparatus of control technology system - Google Patents
For according to control module come the method and apparatus of control technology system Download PDFInfo
- Publication number
- CN108700854A CN108700854A CN201780015870.8A CN201780015870A CN108700854A CN 108700854 A CN108700854 A CN 108700854A CN 201780015870 A CN201780015870 A CN 201780015870A CN 108700854 A CN108700854 A CN 108700854A
- Authority
- CN
- China
- Prior art keywords
- control module
- module
- data
- control
- operation data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 23
- 238000000034 method Methods 0.000 title claims description 33
- 238000012549 training Methods 0.000 claims abstract description 44
- 239000002775 capsule Substances 0.000 claims abstract description 23
- 101100381485 Saccharomyces cerevisiae (strain ATCC 204508 / S288c) BBC1 gene Proteins 0.000 claims abstract description 9
- 238000001514 detection method Methods 0.000 claims abstract description 4
- 238000009826 distribution Methods 0.000 claims description 7
- 230000007613 environmental effect Effects 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000007689 inspection Methods 0.000 claims description 3
- 210000004218 nerve net Anatomy 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000465 moulding Methods 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 239000000567 combustion gas Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000002401 inhibitory effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
- G05B13/029—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion using neural networks and expert systems
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B17/00—Systems involving the use of models or simulators of said systems
- G05B17/02—Systems involving the use of models or simulators of said systems electric
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/0265—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric the criterion being a learning criterion
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B19/00—Programme-control systems
- G05B19/02—Programme-control systems electric
- G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
- G05B19/41885—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by modeling, simulation of the manufacturing system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B2219/00—Program-control systems
- G05B2219/30—Nc systems
- G05B2219/36—Nc in input of data, input key till input tape
- G05B2219/36464—Position, teach, store extreme, full open, closed positions
Landscapes
- Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Manufacturing & Machinery (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Bioethics (AREA)
- General Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Computer Security & Cryptography (AREA)
- Feedback Control In General (AREA)
- Stored Programmes (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Programmable Controllers (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
In order to according to control module(SM, SM1, SM2)Control technology system(TS), receive data capsule(DC, DC1, DC2), in data capsule, there is training structure(TSR)Control module(SM, SM1, SM2)And module information(MTI, MTI1, MTI2)Cross-module block type it is encoded.According to the present invention, according to module type information(MTI, MTI1, MTI2)For technological system(TS)Multiple module types are selected specifically to implement module(EM1, EM2, EM3)In one.In addition, according to module type information(MTI, MTI1, MTI2)By technological system(TS)Operation data channel(BDC)Distribute to control module(SM, SM1, SM2)Input channel(IC).Pass through corresponding operation data channel(BDC)Detection technique system(TS)Operation data(BD)And pass through the input channel for distributing to operation data channel(IC)Send operation data to control module(SM, SM1, SM2).By selected implementation module(EM1, EM2)Implement control module(SM, SM1, SM2), wherein according to training structure(TSR)From the operation data transmitted(BD)Middle export controls data(CD), output control data are for control technology system(TS).
Description
Background technology
In complicated the technological system such as such as control of combustion gas turbine, wind turbine or manufacturing facility, general value
It obtains it is desirable that, carrying out optimization system behavior in view of predetermined standard.For this purpose, in accordance with the epoch control usually using machine learning
Technology.It may be thus possible, for example, to be trained neural network as control module in view of following aspect:In view of multiple standards
Carry out optimisation technique system.
Particularly, during this period, the control of larger facility proposes the safety of control system and flexibility applied
The requirement of raising, the control system have been subjected to the verification process expended in many cases.However this usually makes based on
The application of the control system of habit becomes difficult, because the internal action relationship of control system is generally difficult to from outside to understand simultaneously
And it may be changed according to physical training condition.Additionally, there are the multiple control modules required with different realizations.It is corresponding to this
The application on ground, the control module based on study usually expends ground configuration especially in larger technological system.
Invention content
The task of the present invention is illustrate a kind of method and apparatus for control technology system, the method and described set
The standby more flexible application for allowing the control module based on study.
The task by the method for the feature with Patent right requirement 1, pass through the spy with Patent right requirement 12
The equipment of sign, by the computer program product of the feature with Patent right requirement 13 and by with Patent right requirement
The computer-readable storage medium of 14 feature solves.
In order to, come control technology system, receive data capsule according to control module that is housebroken and/or can training,
In the data capsule, control module and module type information with training structure are compiled in a manner of cross-module block type
Code.Training structure can especially be related to the housebroken, can training, can learning of control module and/or herein in training
Middle constructed structure and/or physical training condition.Training should be particularly understood that basis can be predetermined during the training stage herein
Standard in view of the input parameter for carrying out optimal control module in terms of one or more target parameters mapping(Abbildung).This
The training of sample for example can be by the training of neural network, by analysis model or statistical model recurrence and/or pass through ginseng
The other modes of fitting are counted to realize.It is more to select for the technological system according to the module type information according to the present invention
A module type specifically implements one in the block of mould.In addition, according to the module type information by the fortune of the technological system
The input channel of the control module is given in row data channel allocation.Via technological system described in corresponding operation data Air conduct measurement
Operation data and the input channel by distributing to the operation data channel send the operation data to the control
Molding block.Implement the control module by selected implementation module, wherein according to the training structure from the fortune transmitted
Export control data, the control data are output for controlling the technological system in row data.Control phase can also be exported
The prediction data and/or monitoring data of pass are as control data.
In order to execute according to the method for the present invention, setting control device, computer program product and computer-readable
Storage medium.
An advantage of the invention can be that the control of technological system can be made by specifically implementing module by module type
Structure processed is decoupled from the particular requirement of different control module types as much as possible.This allows entirely different control module more
Simple and more flexible realization.In particular, different control modules can be automatically carried out in same technique system and phase
It can be automatically carried out on different technological systems with control module.
It is specifically distributed by the module type in operation data channel to input channel, utilizes while type can be met operation
Data manipulate the control module.
The advantageous embodiment and expansion scheme of the present invention illustrates in the dependent claims.
Preferably, the control module may include neural network, the recurrence device of data-driven, support vector machines and/or
Decision tree.Above-mentioned implementation can be respectively equipped with training structure and is identified by specific module type information.
A kind of advantageous embodiment according to the present invention, the control module can be with encryptions in the data capsule
Form exist, and be decrypted at least partly by the technological system.It in this way, can be in transimission and storage
When protection control module to prevent unwarranted access.
Preferably, the control module so can be encoded and/or encrypt so that prevent according to the control module
In order to implement the control module and decoded and/or decryption part exports the modular structure of the control module or makes
It becomes difficult.Modular structure can be related to specific layer structure, node structure, networking architecture or the weight of such as neural network
Structure and/or training structure.Control module can be encrypted so herein so that the relevant part energy of only implementation of control module
It is enough to be decrypted by technological system, but specific modular structure keeps maintaining secrecy as much as possible.It therefore, can be only in neural network
Only the calculating routine program that can implement is encoded, enough specific connection of difficulty export are only capable of by the calculating routine program
Web frame.Therefore control module almost can be embodied as flight data recorder.It can be protected in modular structure by this kind of coding or encryption
In include proprietary technology(Know-How).
In addition, the control module can be in the data capsule equipped with digital signature, the digital signature is for example
It can be examined by technological system.Then control module can be implemented according to inspection result.It in this way can be true
Protect the integrality of control module.The establishment, training and/or change of control module can be especially set clearly to be attributed to incumbent institution.
In addition, the data capsule may include the training information about the training of the control module.It is then possible to root
The implementation of the control module and/or the selection for implementing module are carried out according to the training information.Training information especially can be with
It is related to the carried out or subsequent training process of control module.
A kind of advantageous embodiment according to the present invention can give the operation data channel and the input channel point
It Fen Pei not data type, physical dimension, value range and/or subsidiary conditions.It is corresponding giving corresponding operation data channel allocation
Input channel when can examine, whether data type, physical dimension, value range and/or the subsidiary conditions distributed are compatible with.Example
Can such as distribute rice, the second, gram or combination thereof as physical dimension.It can ensure in many cases in this way,
The control module is manipulated with correct operation data.Alternatively, or in addition, it can be tieed up in view of data type, physics
Degree, value range and/or subsidiary conditions check detected operation data, so as to thereby, it is ensured that control module during implementation
Correct manipulation.
Furthermore, it is possible to examine the control number in view of the value range of the control data, value change and/or subsidiary conditions
According to.Because in housebroken control module, the correlation of output data and input data is not usually explicitly known, module
Mistake generally can not be excluded, so error control usually can be avoided by examining predetermined subsidiary conditions.
A kind of advantageous expansion scheme according to the present invention can divide the specific adapter of multiple runtime environments respectively
Dispensing runtime environment.Adapter is used herein to that implementation module is made to adapt to distributed runtime environment.Furthermore, it is possible to detect
The environmental information of runtime environment about the technological system, and can select to be assigned according to the environmental information detected
To the adapter of the runtime environment of the technological system.May then pass through selected adapter makes selected implementation mould
Block is coupled in the runtime environment.In this way, implement module and control module can be as much as possible independently of phase
It is created and realizes to the runtime environment answered.Correspondingly, can be developed as much as possible independently of the type of control module and
Realize adapter.Therefore, control module and runtime environment can be made almost to decouple, thus usually significantly simplifies development process
And/or realize process.
In addition, selected adapter can provide the ability letter of the ability of the runtime environment about the technological system
Breath.The compatibility of the control module and the runtime environment can be examined according to the ability information, and according to described
Compatibility implements the control module.It in this way can be with the automatic realization of simplify control module.
Description of the drawings
The embodiment of the present invention is elaborated below according to attached drawing.Here, respectively in the diagram:
Fig. 1 shows the data capsule according to the present invention with encoded control module,
Fig. 2 shows the technological system with control device according to the present invention,
Fig. 3 shows to export the diagram of control data from operation data by control module, and
Fig. 4 shows the diagram of the interaction of control module and runtime environment.
Specific implementation mode
The root with control module SM that is encoded, housebroken and/or can training is schematically shown in Fig. 1
According to the data capsule DC of the present invention.Control module SM for analogue technique system or technological system a part physics,
On regulation technology and/or random dynamic or other are interrelated.Control module SM may include neural network, data-driven
Recurrence device, support vector machines, decision tree and/or other analysis model or combinations thereof.Control module SM is in data capsule DC
It is encoded to middle cross-module block type, such as with so-called PMML formats(PMML:Predictive Model Markup
Language(Predictive Model Markup Language))Or with proprietary format.
In the present example, control module SM is encrypted with being attached for the data protection in transimission and storage.
In addition, control module SM has training structure TSR.Training structure TSR includes the structure that can learn, preferably in instruction trained in advance
Practice in state.In neural network, networking architecture that training structure TSR can be for example including neuron and between neuron
Connection weight.In the recurrence device of data-driven, training structure TSR can include the coefficient for returning device module.Training knot
Structure TSR can not only be related to the performed training of control module SM but also can be related to its following training.
Data capsule DC includes technology metadata TM and module metadata MM in addition.
In the present example, technology metadata TM includes module type information MTI, training information TI and input and output
Contract dataset IOC.In addition, technology metadata TM include contextual information, such as the creation time of control module SM, about routine
Goal systems information and/or requirement about control module SM to runtime environment information, such as in view of real-time capacity,
Can concurrency, computing resource and/or from the compatibility of different implementation modules.
It is encoded and illustrates to module type information MTI cross-module block types the type of control module SM.For example may be used herein
With explanation, control module SM whether recurrence device based on neural network, based on data-driven, based on support vector machines, based on certainly
Plan tree and/or be based on a combination thereof.Furthermore, it is possible to illustrate control module SM input parameter and/or output parameter and control mould
Other specific requirement, ability and/or characteristics of block SM.
Training information TI describe control module SM carried out and/or subsequent training process and/or physical training condition.
Input and output contract dataset IOC illustrates so-called input and output contract, and the input and output contract is to control module
The behavior of SM proposes subsidiary conditions.By input and output contract dataset IOC can preferably with the format description of cross-module block type it is pre-
Given subsidiary conditions, the value range of such as control module SM, the speed that value changes, value changes and/or input data and/or
The data type of output data.It can be advantageously with the readable format description input and output contract dataset IOC of user, so as to therefore
Ensure the desired behavior of control module SM by checkable mode, wherein the training structure TSR mono- of the control module
As be not that user is readable.
In the present example, module metadata MM includes for example to have created, train and/or changed control module SM's
People and/or one or more digital signature SIG of mechanism.
In addition, module metadata MM can include version information, authority information, about control module SM source and/or
The information of goal systems.In addition, can include about the validity period, about required data in module metadata MM
Process resource and/or about permission or the possible application field for example for monitoring, predicting and/or control explanation.
Fig. 2 is to illustrate to illustrate the technological system with the control device CTL according to the present invention for technological system TS
TS.Technological system TS can be such as power plant, production facility, combustion gas turbine, etc..
Technological system TS possesses the runtime environment RE of control and the data processing for technological system TS.Such operation
When environment, herein be RE may include operating system, high in the clouds/cluster middleware and/or data processing circumstance combination.To this
Example is the Linux clusters with Hadoop/HIVE frames, cluster stream process environment or multinuclear stream process environment.
Control device CTL, such as the control system of combustion gas turbine include all methods for implementing control device CTL
The one or more processors PROC of step and include that module implements system MES.The latter is controlling in the present example
Realized in equipment CTL, but alternatively, or in addition, can at least partly in outside, such as beyond the clouds in realize.Module is real
The system MES of applying can be used as the abstraction layer between control module and runtime environment RE.Module implements system MES packets
Include multiple implementation module EM1, EM2 and EM3 and multiple adapter AD1 and AD2.
It is on technological system or for skill for implementing, installing, initializing and assess to implement module EM1, EM2 and EM3
Art system housebroken and/or the control module that can be trained.It is for controlling mould respectively to implement module EM1, EM2 and EM3
Block type is specific.This kind of implementation module is commonly referred to as interpreter(Interpreter).
Adapter AD1 and AD2 are for making implementation module, and i.e. EM1, EM2 and EM3 adapt to ring when different operations herein
Border.Adapter AD1 and AD2 are specific for runtime environment respectively.
In order to select the specific adapter of runtime environment, module to implement system MES from present in technological system TS
Following environmental information EI is detected in runtime environment RE, the environmental information describes runtime environment RE.According to the ring detected
Border information EI, module implement in system MES selection adapter AD1, AD2 for by being run described in environmental information EI when
Environment is a specific and suitable adapter for RE herein.In the present example, adapter AD2 is certified as fitting
It is selected together in current runtime environment RE and therefore and is coupled on the runtime environment RE.
Selected adapter AD2 then provides the ability information CI about the certain capabilities of runtime environment RE, according to
The ability information can implement system MES to examine the compatibility of control module and runtime environment RE by module.
For control technology system TS, module implements system MES and receives different data capsule DC1 and DC2, the data
Container distinguishes configuration as described in Figure 1.Data capsule DC1 and DC2 separately include specific control module SM1 or
SM2.Preferably, using data capsule DC1 and DC2 as specific messaging to technological system TS.
Control module SM1 and SM2 respectively as described in association with Fig. 1 configuration and be used for analogue technique system
The different physics of system TS's or technological system a part, random on regulation technology and/or or other effects close
Connection.Preferably, control module SM1 and SM2 is determined respectively for the part of the determination of technological system, the adjusting task of determination
Control task and/or the analog type of determination are specific.Control module SM1 and SM2 are encoded to cross-module block type respectively.
By data capsule DC1 and DC2 respectively by for the module type information MTI1 of control module SM1 or SM2 or
Person MTI2 is sent to module and implements system MES.MTI1 and MTI2 illustrates the module type of control module SM1 or SM2 simultaneously respectively
And it can the ground configuration as described in association with Fig. 1 respectively.
After receiving data capsule DC1 and DC2, module implements system MES and unpacks to data capsule and examine respectively
The digital signature of the data capsule.In the case of negative inspection result, inhibit related control module SMl or SM2 into one
Step processing.In addition, module implements system MES for corresponding control module SMl or SM2 according to its technology metadata and root
Examined according to ability information CI, corresponding control module SMl or SM2 and runtime environment RE whether and it is simultaneous with which kind of degree
Hold.According in this, being further processed for corresponding control module SMl or SM2 is carried out.
Encrypted control module SMl and SM2 is decrypted in addition, module implements system MES.Here, it is preferred that only to phase
The relevant part decryption of implementation for the control module answered, although to which control module can be implemented, the module knot of control module
Structure can not be derived with rational consuming.
In addition, module implements system MES according to module type information MTI1 or MTI2 and when necessary according to others
Technology metadata implements module for each control module SMl and SM2 selections for the control module is specific respectively.Current
In embodiment, module EM1 is implemented for control module SMl selections and distribution, and real for control module SM2 selections and distribution
Apply module EM2.Module implement system MES then by selected adapter AD2 by selected implementations module EM1 with
EM2 is coupled on runtime environment RE.
For the operation data for the treatment of technology system TS, module implements system MES and implements control on environment RE at runtime
Module SMl and/or SM2, wherein module implements system MES and entrusts to the implementation of corresponding control module SMl or SM2 point
Implementation the module EM1 or EM2 not distributed.During the implementation of corresponding control module SMl or SM2, according to corresponding control
The input and output contract dataset of molding block SMl or SM2 abide by thus specific input and output contract to monitor and ensure.
Fig. 3 diagrams export control data CD, the control by control module SM from the operation data BD of technological system TS
Molding block is implemented by the adapter AD coupled on technological system TS by implementing module EM.For very clear original
Cause, same technique system TS are schematically shown in the both sides of Fig. 3.Technological system TS, control module SM, implement module EM and
Adapter AD ground configurations preferably as described in association with Fig. 1 and 2.
Control module SM realizes that the neural network has training structure TSR by neural network NN.
Technological system TS possesses the sensor S of the operation data BD for detection technique system TS.Operation data BD can be with
E.g. the physics of technological system TS, on the regulation technology and/or operation parameters caused by structural shape, characteristic, give in advance
Definite value, status data, system data, control data, sensing data and/or measured value.Operation data BD can also especially be wrapped
Include the data for not coming from sensor S.
Operation data BD is detected by the specific operation data channel BDC of technological system TS.Operation data channel BDC
Can be herein specific for the data type of operation data BD, physical dimension, source, function and/or other characteristics.
There is control module SM different input channel IC, the input channel to be assigned to the different of control module SM
Input parameter or input data.Input channel IC can be parameter type, the physics for input parameter or input data
Dimension, source, function and/or other characteristics are specific.
It is allocated IMAP between input channel IC and operation data channel BDC, wherein according to module type information
MTI and operation data channel BDC is distributed respectively to corresponding input channel IC according to input and output contract dataset IOC when necessary
One of.The distribution IMAP implements system MES by module, is carried out preferably by selected implementation module EM.
According to data type, physical dimension, the value range for being respectively allocated to operation data channel BDC and input channel IC
And/or subsidiary conditions, system MES is implemented by module to examine, the data type of input channel IC distributed, physics are tieed up
Degree, value range or subsidiary conditions whether with the data type of the operation data channel BDC distributed, physical dimension, value range or
Subsidiary conditions are compatible with.If it is not the case, then inhibiting the implementation of control module SM.
It regard the operation data BD detected by operation data channel BDC as input number by the input channel IC distributed
According to being conveyed to control module SM.Module implements system MES and implements control module SM by selected implementation module EM, wherein
Control data CD is exported from the operation data BD transmitted according to training structure TSR.Data CD is controlled as control module SM
Output data exported.Control data CD is used herein to control technology system TS and especially can also be that control is relevant pre-
Measured data and monitoring data.
By the specific output channel OC output control data CD of control module SM, the specific output channel is divided
The different output parameters of dispensing control module SM.Output channel OC can be the control data CD for thus exporting respectively
Parameter type, purposes, purpose and/or control function are specific.
Module implements system MES and implements corresponding output channel OC to technology system preferably by selected implementation module EM
The distribution OMAP of one of multiple control channel CDC of system.The distribution is herein according to module type information MTI and when necessary
It is carried out according to input and output contract dataset IOC.It is examined when distributing OMAP, control channel CDC and output channel OC institute
Whether the data type of distribution, physical dimension, value range etc. are compatible.If it is not, then inhibiting the reality of control module SM
It applies.
During the implementation of control module SM, system MES is implemented by module and is monitored according to input and output contract dataset IOC
With ensure to abide by related input and output contract.
Fig. 4 illustrates the mutual of different control module SM1 ..., SMM and different runtime environment RE1 ..., REN
Effect.
Runtime environment RE1 ..., REN respectively by the specific multiple adapter AD1 of runtime environment ... or AND
One of them is coupled to module as described above implements on system MES.In addition, control module SM1 ..., SMM difference
By the specific multiple implementation module EM1 of module type ... or one of EMM is coupled to module as described above
On implementation system MES.
Operation data environment RE1 at runtime ... or arrival in one of REN or pass through one of these runtime environments
The detection of the operation data is promoted:By the runtime environment RE1 being detected ... or the processing of REN via operation
When the specific adapter AD1 of environment ... or ADN and module type specifically implement module EM1 ... or EMM is to specific control
Molding block SMl ... or the commission DBD of SMM carries out the processing to the data-driven of operation data.The commission DBD leads to herein
It crosses module and implements system MES to facilitate.
By corresponding control module SMl ... or the output of the derived control data of SMM promotes:The control of technological system
Process implements the commission DCD that system MES is facilitated by module.With from corresponding control module SMl ... or SMM is via phase
The module type distributed answered specifically implements module EM1 ... or EMM and the specific adapter of corresponding runtime environment
AD1 ... or AND to the runtime environment RE1 that is distributed ... or the mode of REN carries out commission DCD, the commission according to
It controls data CD and controls the technological system TS.
Claims (14)
1. a kind of for according to control module that is housebroken and/or can training(SM, SM1, SM2)Carry out control technology system
(TS)Method, wherein by processor(PROC)
a)Receive data capsule(DC, DC1, DC2), in the data capsule, there is training structure(TSR)Control module
(SM, SM1, SM2)And module type information(MTI, MTI1, MTI2)It is encoded in a manner of cross-module block type,
b)According to the module type information(MTI, MTI1, MTI2)For the technological system(TS)Select multiple module types special
Fixed implementation module(EM1, EM2, EM3)In one,
c)According to the module type information(MTI, MTI1, MTI2)By the technological system(TS)Operation data channel
(BDC)Distribute to the control module(SM, SM1, SM2)Input channel(IC),
d)Via corresponding operation data channel(BDC)Detect the technological system(TS)Operation data(BD)And by dividing
The input channel in operation data channel described in dispensing(IC)Send the operation data to the control module(SM, SM1,
SM2),
e)By the selected implementation module(EM1, EM2)Implement the control module(SM, SM1, SM2), wherein according to
The training structure(TSR)From the operation data transmitted(BD)Middle export controls data(CD), and
f)Export the control data(CD)For controlling the technological system(TS).
2. according to the method described in claim 1, it is characterized in that, the control module(SM, SM1, SM2)Including nerve net
Recurrence device, support vector machines and/or the decision tree of network, data-driven.
3. method according to any one of the preceding claims, which is characterized in that the control module(SM, SM1, SM2)
In the data capsule(DC, DC1, DC2)In exist in an encrypted form, and pass through the technological system(TS)At least partly
Ground is decrypted.
4. method according to any one of the preceding claims, which is characterized in that the control module(SM, SM1, SM2)
It is so encoded and/or is encrypted so that prevented according to the control module to implement the control module(SM, SM1,
SM2)And decoded and/or decryption part exports the modular structure of the control module or it is made to become difficult.
5. method according to any one of the preceding claims, which is characterized in that the control module(SM, SM1, SM2)
In the data capsule(DC, DC1, DC2)In equipped with digital signature(SIG), check the digital signature(SIG), and root
Implement the control module according to inspection result.
6. method according to any one of the preceding claims, which is characterized in that the data capsule(DC, DC1, DC2)
Including about the control module(SM, SM1, SM2)Training training information(TI), and according to the training information(TI)
Carry out the control module(SM, SM1, SM2)Implementation and/or the implementation module(EM1, EM2)Selection.
7. method according to any one of the preceding claims, which is characterized in that give the operation data channel(BDC)With
The input channel(IC)Distribute data type, physical dimension, value range and/or subsidiary conditions respectively, and will be corresponding
Operation data channel(BDC)Distribute to corresponding input channel(IC)When examine, the data type distributed, the physics
Whether dimension, described value range and/or the subsidiary conditions are compatible with.
8. method according to any one of the preceding claims, which is characterized in that in view of the control data(CD)Value
Range, value change and/or subsidiary conditions examine the control data(CD).
9. method according to any one of the preceding claims, which is characterized in that specifically fit multiple runtime environments
Orchestration(AD, AD1, AD2)It is respectively allocated to runtime environment(RE), for making implementation module(EM1, EM2, EM3)Adapt to institute
The runtime environment of distribution(RE),
Detection is about the technological system(TS)Runtime environment(RE)Environmental information(EI),
According to the environmental information detected(EI)To select to be assigned to the technological system(TS)The operation when ring
Border(RE)Adapter(AD2),
Pass through the selected adapter(AD2)Make the selected implementation module(EM1, EM2)It is coupled to the operation
When environment(RE)On.
10. according to the method described in claim 9, it is characterized in that, the selected adapter(AD2)It provides about described
Technological system(TS)The runtime environment(RE)Ability ability information(CI), and
According to the ability information(CI)Examine the control module(SM, SM1, SM2)With the runtime environment(RE)It is simultaneous
Capacitive, and the control module is implemented according to the compatibility.
11. method according to any one of the preceding claims, which is characterized in that according to the operation data transmitted
(BD)Regularly retraining or the constantly trained control module(SM, SM1, SM2).
12. a kind of for according to control module that is housebroken and/or can training(SM, SM1, SM2)Carry out control technology system
(TS)Equipment, the equipment is configured for implementing method according to any one of the preceding claims.
13. a kind of computer program product, the computer program product is configured for implementing according in claim 1 to 11
Any one of them method.
14. a kind of computer-readable storage medium, the computer-readable storage medium has according to claim 13 institute
The computer program product stated.
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE102016203855.9 | 2016-03-09 | ||
DE102016203855.9A DE102016203855B4 (en) | 2016-03-09 | 2016-03-09 | Method and device for controlling a technical system using control models |
PCT/EP2017/052133 WO2017153095A1 (en) | 2016-03-09 | 2017-02-01 | Method and device for controlling a technical system by means of control models |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108700854A true CN108700854A (en) | 2018-10-23 |
CN108700854B CN108700854B (en) | 2022-03-01 |
Family
ID=57984918
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201780015870.8A Active CN108700854B (en) | 2016-03-09 | 2017-02-01 | Method and device for controlling a technical system according to a control module |
Country Status (8)
Country | Link |
---|---|
US (1) | US11269297B2 (en) |
EP (1) | EP3400490B1 (en) |
JP (1) | JP6688902B2 (en) |
KR (1) | KR102222296B1 (en) |
CN (1) | CN108700854B (en) |
DE (1) | DE102016203855B4 (en) |
ES (1) | ES2947032T3 (en) |
WO (1) | WO2017153095A1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102016203855B4 (en) * | 2016-03-09 | 2023-11-02 | Siemens Aktiengesellschaft | Method and device for controlling a technical system using control models |
KR102186009B1 (en) | 2019-03-08 | 2020-12-04 | 한국전자통신연구원 | System and method for generating secure profile of container instance |
EP3889710A1 (en) * | 2020-03-30 | 2021-10-06 | Siemens Aktiengesellschaft | Control system of a technical assembly |
JP7342833B2 (en) * | 2020-10-16 | 2023-09-12 | 横河電機株式会社 | Control device, controller, control system, control method, and control program |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101960397A (en) * | 2008-03-03 | 2011-01-26 | 阿尔斯托姆科技有限公司 | Fuzzy logic control and optimization system |
US20140214733A1 (en) * | 2013-01-30 | 2014-07-31 | Siemens Aktiengesellschaft | Method And Apparatus For Deriving Diagnostic Data About A Technical System |
CN104085265A (en) * | 2014-06-12 | 2014-10-08 | 江苏大学 | Adaptive offline neural network inverse-control system and method for energy-regenerative suspension |
CN104142254A (en) * | 2014-08-06 | 2014-11-12 | 浙江大学 | Fault diagnosis method for flow passage of steam turbine |
CN104537435A (en) * | 2014-12-18 | 2015-04-22 | 国家电网公司 | Distributed power source optimizing configuration method based on user-side economic indexes |
CN104573170A (en) * | 2014-10-13 | 2015-04-29 | 芜湖扬宇机电技术开发有限公司 | Method for determining initial pressure of optimal operation of steam turbine |
CN104821764A (en) * | 2015-04-23 | 2015-08-05 | 王金全 | Pulse load power equivalent method according to system dynamic characteristics |
CN104836498A (en) * | 2015-04-21 | 2015-08-12 | 苏州德锐朗智能科技有限公司 | Auto-tuning PID generator control system based on artificial neural network |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE19841165A1 (en) | 1998-09-09 | 2000-03-16 | Abb Research Ltd | Process data validation model determination in technical plant, involves controlling model fine tuning remotely by model selection- and optimization module connected over proxy-/adapter module with validation tool |
JP4017272B2 (en) | 1998-12-16 | 2007-12-05 | 中国電力株式会社 | Plant state estimation / prediction apparatus and method |
KR100401801B1 (en) * | 2001-03-27 | 2003-10-17 | (주)텔레시스테크놀로지 | Orthogonal frequency division multiplexing/modulation communication system for improving ability of data transmission and method thereof |
JP2005293109A (en) * | 2004-03-31 | 2005-10-20 | Canon Inc | Software execution management device, software execution management method, and control program |
US7738975B2 (en) * | 2005-10-04 | 2010-06-15 | Fisher-Rosemount Systems, Inc. | Analytical server integrated in a process control network |
US7599750B2 (en) | 2005-12-21 | 2009-10-06 | Pegasus Technologies, Inc. | Model based sequential optimization of a single or multiple power generating units |
EP1979794A4 (en) | 2006-02-03 | 2010-03-17 | Rech 2000 Inc | Intelligent monitoring system and method for building predictive models and detecting anomalies |
DE102007001024B4 (en) | 2007-01-02 | 2008-10-02 | Siemens Ag | Method for computer-aided control and / or control of a technical system, in particular of a gas turbine |
US8127271B2 (en) * | 2007-03-02 | 2012-02-28 | International Business Machines Corporation | Method and system for accessing a resource implemented in a computer network |
EP2891997A1 (en) * | 2014-01-06 | 2015-07-08 | Fujitsu Limited | Methods and apparatus for including a confidential structural component in a third party remote product simulation |
US9684490B2 (en) * | 2015-10-27 | 2017-06-20 | Oracle Financial Services Software Limited | Uniform interface specification for interacting with and executing models in a variety of runtime environments |
DE102016203855B4 (en) * | 2016-03-09 | 2023-11-02 | Siemens Aktiengesellschaft | Method and device for controlling a technical system using control models |
-
2016
- 2016-03-09 DE DE102016203855.9A patent/DE102016203855B4/en active Active
-
2017
- 2017-02-01 JP JP2018547436A patent/JP6688902B2/en active Active
- 2017-02-01 US US16/082,002 patent/US11269297B2/en active Active
- 2017-02-01 CN CN201780015870.8A patent/CN108700854B/en active Active
- 2017-02-01 EP EP17703704.1A patent/EP3400490B1/en active Active
- 2017-02-01 ES ES17703704T patent/ES2947032T3/en active Active
- 2017-02-01 KR KR1020187029129A patent/KR102222296B1/en active IP Right Grant
- 2017-02-01 WO PCT/EP2017/052133 patent/WO2017153095A1/en active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101960397A (en) * | 2008-03-03 | 2011-01-26 | 阿尔斯托姆科技有限公司 | Fuzzy logic control and optimization system |
US20140214733A1 (en) * | 2013-01-30 | 2014-07-31 | Siemens Aktiengesellschaft | Method And Apparatus For Deriving Diagnostic Data About A Technical System |
CN104085265A (en) * | 2014-06-12 | 2014-10-08 | 江苏大学 | Adaptive offline neural network inverse-control system and method for energy-regenerative suspension |
CN104142254A (en) * | 2014-08-06 | 2014-11-12 | 浙江大学 | Fault diagnosis method for flow passage of steam turbine |
CN104573170A (en) * | 2014-10-13 | 2015-04-29 | 芜湖扬宇机电技术开发有限公司 | Method for determining initial pressure of optimal operation of steam turbine |
CN104537435A (en) * | 2014-12-18 | 2015-04-22 | 国家电网公司 | Distributed power source optimizing configuration method based on user-side economic indexes |
CN104836498A (en) * | 2015-04-21 | 2015-08-12 | 苏州德锐朗智能科技有限公司 | Auto-tuning PID generator control system based on artificial neural network |
CN104821764A (en) * | 2015-04-23 | 2015-08-05 | 王金全 | Pulse load power equivalent method according to system dynamic characteristics |
Also Published As
Publication number | Publication date |
---|---|
DE102016203855A1 (en) | 2017-09-28 |
DE102016203855B4 (en) | 2023-11-02 |
EP3400490B1 (en) | 2023-03-29 |
CN108700854B (en) | 2022-03-01 |
JP2019510310A (en) | 2019-04-11 |
EP3400490A1 (en) | 2018-11-14 |
US11269297B2 (en) | 2022-03-08 |
JP6688902B2 (en) | 2020-04-28 |
US20200301375A1 (en) | 2020-09-24 |
KR20180125156A (en) | 2018-11-22 |
WO2017153095A1 (en) | 2017-09-14 |
KR102222296B1 (en) | 2021-03-02 |
ES2947032T3 (en) | 2023-07-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108700854A (en) | For according to control module come the method and apparatus of control technology system | |
US10037443B2 (en) | Industrial simulation using redirected I/O module configurations | |
US11694212B2 (en) | Decentralized governance regulatory compliance (D-GRC) controller | |
EP3422639A2 (en) | Performance testing method and apparatus for an industrial system deployed on cloud | |
CN108133279B (en) | Wind power probability prediction method, storage medium and device | |
EP3002649A2 (en) | Industrial simulation using redirected i/o module configurations | |
WO2021007006A1 (en) | Dual machine learning pipelines for transforming data and optimizing data transformation | |
US11550310B2 (en) | Industrial control system data tap and model for managing industrial control system | |
US20210365762A1 (en) | Detecting behavior patterns utilizing machine learning model trained with multi-modal time series analysis of diagnostic data | |
US10747207B2 (en) | System and method for accurate automatic determination of “alarm-operator action” linkage for operator assessment and alarm guidance using custom graphics and control charts | |
CN112904807B (en) | Industrial analysis system, method, and non-transitory computer readable medium | |
US20200367834A1 (en) | Device for predicting body weight of a person and device and method for health management | |
CN115037790A (en) | Abnormal registration identification method, device, equipment and storage medium | |
AU2021218217A1 (en) | Systems and methods for preventative monitoring using AI learning of outcomes and responses from previous experience. | |
Steimers et al. | Sources of risk and design principles of trustworthy artificial intelligence | |
GB2615243A (en) | Systems and methods for micro-credential accreditation | |
CN116529034A (en) | Generating digital twins, methods, systems, and computer program products | |
US20240160198A1 (en) | Industrial data source writeback | |
EP4369121A1 (en) | Industrial data extraction | |
EP4369120A1 (en) | Industrial automation data quality and analysis | |
US20240160193A1 (en) | Industrial automation data staging and transformation | |
CN117593263A (en) | Image tampering detection method, device, storage medium and electronic equipment | |
KR20240105082A (en) | Method and Apparatus for Detecting Network Attack and Intrusion based on Machine Learning | |
CN118095803A (en) | Logistics resource integration and scheduling platform and method based on big data | |
CN117254955A (en) | Network safety sensing system of electric power system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |